I often write about the need to use evidence to inform the decisions that we make in schools. However, this is not an easy path. Evidence can be ambiguous. The question that you want answered may not have been directly addressed and, even if it has, the results may be inconclusive. A new initiative may be extremely appealing but there may be very little research on it. Basing your approach on evidence can seem cautious and slow.
And some of the world’s most successful organisations don’t work this way. Big businesses and silicon valley start-ups are famous for innovating. They are busy building the future rather than tortuously analysing research from past. Perhaps this is a better model for education to follow.
There are two reasons why we should be wary of this argument.
1. Innovations are not always improvements
The originators of the ‘whole language’ approach to teaching reading had good intentions. They saw what they thought were very old-fashioned methods where children were drilled in letter-sound relationships. They felt this sucked all the joy out of learning to read and didn’t sit well with cutting-edge constructivist learning theory. You learn to read for a purpose, they reasoned, and so learning to read needed to be structured around the reading of real books. You can’t do that if you are initially restricted to simple letter-sound correspondences and so children needed other strategies to figure out words. Proponents looked to what skilled readers do and they suggested that skilled readers inferred words from context. Students could be supported to predict words by using books with text structures such as a repeating refrain. Instead of forcing children through a sequence of exercises, specific teaching points should be addressed as they naturally arise in the context of engaging texts.
We now know that this innovative approach was flawed. Three government reports from the U.S., U.K. and Australia all evaluated the evidence and showed that a bottom-up approach to teaching reading, starting with the key building blocks of phonics, is the most effective way of enabling the greatest number of children to learn to read. Moreover, this teaching needs to be systematic; the learning has to follow a logical and structured sequence rather than relying on teaching features as they happen to appear within texts. These days, few former advocates of whole language still argue for the 1980s version, stressing instead that they favour a ‘balanced’ approach which they claim includes more of an emphasis on phonics. This is debatable but it illustrates that a once fashionable and obvious innovation has been disowned by pretty much everyone.
We don’t know how many students were failed by whole language. Many children will learn to read regardless of the method but it seems likely that the trend for whole language caused real harm and contributed to the development of reading difficulties in many young people.
2. Businesses fail
If a private company had introduced a similar innovation to whole language then it would likely have gone bust*. This is the logic of markets. Businesses are able to innovate but when they do so, they take on risk. The free market evolutionary principle is fine with that: businesses are supposed to rise and fall with the invisible hand of the market generating gradual improvement.
A couple of years ago a shop opened up in a town near me. It was innovative. It sold home-brewing equipment but instead of the slightly 1970s vibe you get in many home-brew shops, this had a hipster feel. They also sold equipment for making sausages so it was an eclectic mix of concepts. It was a fascinating little shop – there was nothing else like it around. And now it has gone.
People don’t tend to think that schools that adopt flawed approaches should be allowed to fail and disappear. But this seems a necessary element of a system that relies on innovation rather than research. If we are going to continue to put taxpayer money into schools, regardless of their successes and failures, then we should expect them to make decisions on the basis of the best available evidence, rather than, “Oh, look! A shiny thing!”
It is also worth noting that there are other models of school provision that sit well with the idea of innovation. Journalists tend to delight in writing reports about U.S. Charter Schools or U.K Free Schools that have failed, implying that these failures demonstrate the failure of the broader policy. But failure is the point! These policies are intended to open up the field to innovation so that effective schools will prosper and ineffective ones will fail.
Show me the evidence
I am yet to be convinced about the Charter/Free School model. These systems are designed to generate diversity and so there are necessarily some schools that I approve of more than others. And clearly, there are some that simply could not exist under the old system. Yet where the old system of centrally managed and funded schools is still largely still in place, the moral imperative is not to roll a dice and innovate; it is to use the best evidence available, however complicated and boring that may be.
*a notable exception to this rule were the innovative banks deemed too big to fail during the global financial crisis – a model worth avoiding
Filling the pail 
Reblogged this on The Echo Chamber.
Interesting article but hasn’t the very nature of schools changed? So looking at evidence of success from the 1940s or 1950s isn’t very helpful as students could be ejected from the school system at their maximum achievable level and find a job, not really true today.
Can we look at anything pre-2000 when we didn’t have the internet and the illusion/reality of limitless facts (not knowledge or expertise) has changed what society thinks we should teach? I know you have debated this in the past.
Neuroscience has great evidence (do something, image the brain, see if it repeats), your article on why that is worrisome was great. So maybe understanding the minute detail of the brain isn’t helpful.
Maybe we need a new idea because we live in a new age. Can we support it with empirical evidence, probably not, at least not right now. But we do need to experiment because we can’t say we have got it right yet.
School leaving age has nothing to do with the process-product studies of the 1960s and 1970s because the same students were tracked from start to end then performance was measured. Similarly, it would an odd RCT that had students leaving school midway through.
The internet doesn’t change anything much because prior to its advent people made exactly the same arguments about books and TV. In my view, the best piece on why you can’t just look things up is still this one:
Click to access LookItUpSpring2000.pdf
Even if we did accept the implausible idea that students’ brains are somehow different to how they were in the 1960s then on what basis would we proceed? How do we know our new and innovative approach will be better and not worse?
Student brains are not different these days but the availability, analysis, presentation and storage of information is very different.
Learning is about exposure to information and how people process that information, both.
The before internet there were books argument is a little feeble and there is no doubt whatsoever that the invention of books did indeed change learning on a global scale. And yes, people do keep books on shelves and in libraries, they do not commit everything to memory. Shock horror…people refer to reference books when they cant recall every detail or they wish to research.
It surprises me how often debate around such issues suddenly turns into “you cant just look everything up”, a position which really doesn’t need an rct to confirm.
Ps….most teachers innovate most days, whatever system they work in
Interesting article but hasn’t the very nature of schools changed? So looking at evidence of success from the 1940s or 1950s
Even if this is true: so what? It does not effect the argument that innovative schools have to be allowed to fail, or that we should pay attention to such evidence as we have.
It is perhaps an argument for more innovation, and therefore for more free schools. Fortunately, the government seems to want to have them.
It’s not true anyway. The school rooms of the 1940s and 1950s were not that different. Teachers taught lessons in classrooms of 10 to 30 students, with mostly the same subjects on offer. In Maths the curriculum hasn’t even changed that much. A Maths teacher transplanted from the 1950s to my classroom would not be fine, once they learned not to hit the kids.
This concept that things are rapidly evolving stems from our excitement over changes in IT and its flow on effects. In every other field the world changes very slowly. Even science isn’t that fast moving — the last really big revolution in Physics was well over a century ago. The Chemistry lessons of today are barely changed from when I was a student, because Chemistry isn’t that fast moving anyway.
The idea that the world changes quickly is largely a myth. What was true 100 years ago tends to be true now.
Pingback: Beware the innovators and changemakers | Le Soutien au Comportement Positif (SCP) au Canada
This internet changes everything argument is another form of bogus claims about 21st century skills and jobs. Guess what it is 2016. Every teacher who has worked more than 15 years has been working in a 21st century job for quite some time. Most have the internet at hand via a smartphone or a laptop.
How many look things up during their typical work day? How many have searchable access to the last Professional Development slide decks and latest research journals that they refer to several times a day?
How many use data driven feedback on their students progress on a daily basis?
How many think new teachers should be given a website with links to the relevant knowledge and skip all the course work beyond high school that goes into becoming a teacher?
Alternatively how many think you can’t teach a subject without rapid recall of all the relevant knowledge on the subject and how to impart it to a possibly unreceptive audience? Can you imagine a math teacher who didn’t know any calculus because he could just look it up teaching the subject or an English teacher taking on teaching literary criticism armed with the ability to search the internet and no prior knowledge of the subject.
Unless the “you can look it up argument so don’t need to know it” applies in a heavily knowledge intensive field such as teaching it is unlikely to apply anywhere else.
(Of course there are some new things to teach about the internet verses a library – that the fiction and non-fiction sections have not been well organized for example. But that people think these things need teaching rather than kids just looking up how to sort fiction from fact via google again illustrates the point some facts need to be taught.)
I suppose it depends on how you view the use of the technology. A good day with my students has me looking up things all the time as they are asking questions way outside of what could be required by curriculum knowledge (think the last one involved commensalistic relationships in a mudflat, not something I know offhand). But we are a 1:1 school so much of our resources is digital.
Would be great if we had better data feedback on students, nobody seems to be working on this (most LMS seem terrible at providing useful data for teaching) was wistful for the google glass to interface this, but alas.
Agree completely that we shouldn’t use technology for recall of core literacy and numeracy skills, this is what learning is about. This technology isn’t only about facts, it can provide the means to develop new skills (many of which are needed in today’s business) provide authentic audiences (something that can make work incredibly valuable to students) and provide social frameworks giving students the ability to discuss many of the ideas being taught.
My comment was aimed at those claiming no one needs any knowledge because they can look it up. I am doubtful they find this true for themselves or their colleagues.
I had to sit through a ‘Twenty-First Century Skills’ presentation this week. It is a flawed argument and as with so many derives from the use of a top-down model of education as a nationally-strategic asset, rather than a bottom-up model of education as personal need.
Education only drives national assets *by* changing individual lives. But the human needs behind those lives most likely have not changed very much for millennia – hence as you say, there is nothing inherently valuable in jettisoning things that have worked in the past – even if we don’t know why. Nor is it tenable to redefine ‘what works’ to align it with a particular philosophy.
And as for the internet – Google may provide facts but it cannot develop wisdom.
Pingback: Skepticism Improves Innovation - Learning and the Brain blogLearning and the Brain blog