Not basic

Embed from Getty Images

Project Follow Through was set up in the late 1960s and represents the largest educational experiment that has ever been conducted. It followed a ‘horse race’ design. Different groups of researchers developed programmes for early years education and these were then trialled with disadvantaged children. The concept was one of ‘planned variation’ and the aim was to find which programmes were the most effective.

Due to its size, the study was necessarily messy. So it does not suit an experimental purist and the results can, if you wish, be explained away. However, it did have one key element in its favour, what we might call ‘active controls’. In other words, different researchers were all motivated to demonstrate that their programme worked and they had resources available to give it their best shot. By contrast, much experimental work in education involves comparing a project that researchers are excited about with a business-as-usual approach that is hardly going to be as exciting and shiny as the new intervention. Any positive results may therefore be as a result of a placebo or hawthorne effect. This is why I have called for ABC randomised controlled trials where two competing interventions are compared with each other and a control, but I digress.

There was one clear winner of Project Follow Through – ‘Direct Instruction’ developed by Siegfried Engelmann and colleagues. Direct Instruction was characterised as a ‘basic skills’ approach in the Project Follow Through research because it focused on the acquisition of literacy and numeracy. Other, less effective, programmes were classed ‘affective’  or ‘cognitive’. The former focused on how children felt and their self-esteem, whereas the latter focused on supposedly ‘higher order’ skills such as problem-solving.

This way of classifying educational objectives is entirely wrong and deeply misleading.

David C Geary has proposed that we should think of two broad categories of abilities. The first category is ‘biologically primary’ knowledge and skills. We are not born with the ability to speak our native language but we do have an ability to learn how to do this with seemingly little effort. Similarly, we can learn basic maths skills in this way – children in societies across the world tend to count objects when playing.

These abilities provide a survival benefit and so have been shaped over many hundreds of thousands of years by the process of evolution. We are hard wired to not only learn this knowledge, but we are intrinsically motivated to do so. Learning through the process of play is enjoyable.

In contrast, ‘biologically secondary’ knowledge and skills arrived late in evolutionary terms. For most of human history, nobody read, wrote or completed even basic calculations, and so these abilities have not been acted on by evolution. Although qualitatively different to biologically primary knowledge and skills, there is a clear relationship in that biologically secondary abilities work by hacking biologically primary ones. Reading and writing sit on the back of speaking and listening abilities and mathematics deploys basic numerosity skills.

We are not intrinsically motivated to learn biologically secondary knowledge and skills and so we need to develop them through an effortful process of deliberate practice. Attempts to find a teaching method that will make learning to read as effortless as learning to speak are therefore likely to fail because they are based on a misunderstanding.

In this view, literacy and numeracy are not basic at all. They are, instead, sophisticated cultural products. It is these abilities that are ‘higher order’.

They also tend to be quite subject specific. General problems solving strategies such as ‘means ends analysis’ are biologically primary, whereas secondary problem solving strategies seem to be associated with specific classes of problems such as algebra or plumbing. This leads to the question that crops up most often in education: Why am I learning X if I will never need X in real life?

This is the question that motivates us to attempt to develop generic skills such as problem solving, collaboration or critical thinking. However, if there were such generic strategies that applied to many different contexts then it seems likely that they would confer an evolutionary advantage and so why have we not evolved to acquire these strategies alongside the rest of our biologically primary knowledge?

When we attempt to teach general problem solving, collaboration or critical thinking strategies, we are probably teaching biologically secondary, domain specific knowledge. We may think we are developing children’s collaborative skills but we are actually teaching children how to collaborate in a way that meets the norms and expectations of Mr Jones’s maths class. It is likely that little will transfer to the real-life situations that it is intended to support. It may be better to close schools and send kids out directly into society if we want them to learn norms that will be of practical use.

This makes sense if we have no clear idea of what an academic education is for. I believe there is a moral aspect to passing on these cultural fruits; that they are a good in their own right. However, there is also the possibility that academic knowledge succeeds where attempts to teaching generic skills fail in that it truly does offer us something that is transferable.

If we want students to ‘learn how to learn’ then the most effective method is to teach them to read because they can then learn through reading. But mastering the mechanic of reading, although vital, is just the start of learning to read. In order to comprehend text, we need rich background knowledge, the sort of knowledge of history, geography, science and literature that schools are designed to deliver.

learning this stuff is worth the effort, but it is definitely not basic.


8 thoughts on “Not basic

  1. Tom Burkard says:

    RCTs will inevitably be ineffective if you are evaluating anything which conflicts with teachers’ training, which is why we need interventions like Follow Through and Clackmannanshire. The latter, at least, has had a profound effect on policy in England, even if relatively few primary schools have completely abandoned the disastrous ‘searchlights’. Although Kenneth Goodman and Frank Smith are now oddities that progressive educators would just as soon forget, the early reading pedagogy advocated in the EEF toolkit looks suspiciously like the 1998 National Literacy Strategy, which was little more than a truce cobbled together to salvage as much of the whole language project as possible. For instance, the ‘simple view of reading’–i.e., when pupils have good decoding skills, ‘reading comprehension’ closely tracks oral language comprehension–is ignored. It’s hard to take the EEF seriously when you consider that there is almost no evidence that teaching children to understand one text will transfer to other texts.

    Another objection to RCT: unlike trials of drugs, which are relatively cheap and unproblematic, it is both expensive and difficult to conduct RCTs in education. To some extent, this gives the educational establishment a veto over any intervention they don’t like.

  2. The Quirky Teacher says:

    Hi, great post, but I wonder if I could convince you to question the shibboleth that is ‘biologically primary knowledge’? When I look at the research, none of it seems to take into account the intensity, quantity or consistency of mother-child interactions and I think that perhaps this idea that spoken language is ‘easy’ to acquire and therefore doesn’t need much in the way of direction is wrong. Firstly, so many disadvantaged children start school these days unable to speak, let alone have a conversation. When I talk the mothers about this, they say ‘Oh he doesn’t like talking, so we let him be’. Also, if you think about your own experiences as a parent (observing your own child), learning to talk is also born out of the extreme frustration of not being able to communicate needs and wants which then propels the child to put in more effort (or just continue having tantrums if the parent allows it) to speak. So, I tend to conclude that, away from the researchers, mothers mostly make very deliberate attempts to teach their child to speak, model sentences with new vocabulary and expect her child to repeat them back, expose them to every day conversation too – this is the teaching part. The practice is arduous for the child because he is frustrated and (hopefully) the mother doesn’t allow tantrums, and then there is the equivalent of retrieval practice every time a mother and child sit together to eat, with simple conversations about what we shall do and what we have done. Nobody seems to acknowledge this deliberate teaching and practice; they just call it ‘primary knowledge’.

    • I suppose it’s not quite right to suggest that biologically primary knowledge is ‘easy’ to acquire. Biologically primary knowledge is knowledge we have evolved to acquire from the environment without explicit teaching. We have mental modules of some kind that are primed to acquire it and so it does not pass through working memory and is therefore not subject to those constraints. However, you still need the right environment to interact with. Some children will also have impairments that prevent them acquiring biologically primary knowledge and, in some circumstances, explicitly teaching this knowledge may be an effective intervention.

      • It’s a model that makes useful predictions. That’s all. Nobody can point to neurological analogues or anything like that. If you have a critique that is a little more developed than ‘hmmm’ then I would be interested to read it.

      • The Quirky Teacher says:

        Fair point. I only did a small amount of neuroscience in my biology degree, but I’m struggling to see how there would be magical neural pathways that bypass working memory – even if there was, how would we know? Can we ask a baby whether they’re having to think very hard about what their mother is saying to them? I’m thinking that working memory works in exactly the same way for a 1 year old as it does for a 2 year old and so forth – it’s just that the baby has so little else to distract because he is primed to look at and want to be with his mother most of the time. Maybe this information doesn’t somehow bypass his working memory; rather, he is the most perfect ‘student’ because he dotes on his mother so much.

      • It is not a neuroscience theory. It is a cognitive theory. I am pointing out that we don’t need to be able to point to neuroscience mechanisms to justify a cognitive theory. We cannot know what a baby is thinking but we can test some of the predictions of Geary’s theory. There is plenty of debate around this.

Leave a Reply

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

You are commenting using your 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

This site uses Akismet to reduce spam. Learn how your comment data is processed.