Skills vs knowledge, 13 years on
What can we learn from widespread dissatisfaction with the Curriculum for Excellence?
Back in 2010, Scotland introduced a new curriculum - the Curriculum for Excellence. The Curriculum for Excellence explicitly reduced the content knowledge in the curriculum, and organised the curriculum around a set of content-free skills statements like: “Using what I know about the features of different types of texts, I can find, select and sort information from a variety of sources and use this for different purposes.”
Yesterday, the Sunday Times had a long article on the curriculum’s performance so far. It does not make for great reading. They quote parents unhappy with low standards and teachers unhappy with vague and unhelpful documentation. Most remarkably, they quote one of the architects of the curriculum, Keir Bloomer, accepting that the curriculum may have gone too far in reducing its emphasis on knowledge.
“The problem is we did not make sufficiently clear that skills are the accumulation of knowledge. Without knowledge there can be no skills.”
The skills-knowledge debate is one of the perennial educational debates in many countries. If, after more than a decade, this flagship skills-based curriculum is not working out, then that is of relevance for educators globally and is worth exploring further.
The skills vs knowledge debate
When I taught English, my aim wasn’t for my students to leave school able to recite some disconnected facts. I wanted them to be able to read critically and infer deeper meanings. I wanted them to be able to write for a range of purposes. I wanted them to be able to speak eloquently and understand spoken debates. To that extent, I’d have agreed that the statements in the Curriculum for Excellence are not a bad guide to the end goals of education.
However, when it comes to teaching practice, the question is not whether these statements are legitimate goals or not. The question is whether or not they are helpful guides for teachers who would like their students to achieve them.
When you dig into the research, you find that the best way to teach these complex skills is by breaking them down into smaller parts, teaching those smaller parts, and gradually combining together the constituent parts in increasing complexity. This is essentially because our working memories are very limited, and struggle to handle lots of new items of information at once. We can’t learn complex skills directly.
What you also find is that in most school subjects, those smaller parts look like what we typically call 'knowledge'. For example, in order for students to “find, select and sort information from a variety of sources” they need to know the meanings of some words. In order to “describe the main features of conflicting world belief systems in the past and can present informed views on the consequences of such conflict for societies then and since”, they need to know when different historical events happened.
In the words of Dan Willingham, Professor of Cognitive Psychology at the University of Virginia:
“Data from the last 40 years lead to a conclusion that is not scientifically challengeable: thinking well requires knowing facts, and that's true not simply because you need something to think about. The very processes that teachers care about most – critical thinking processes such as reasoning and problem solving – are intimately intertwined with factual knowledge that is stored in long-term memory (not just found in the environment).”
So what is the right balance between knowledge and skills?
This is the wrong question. If you are asking this question, you will only get bad answers.
Asking what the right balance is between knowledge and skills is like asking what the right balance is between ingredients and cake.
The ingredients make the cake, just as the knowledge makes the skill.
Knowledge and skills are not a pendulum. You can't think of this debate as one where the pendulum tipped too far towards skills, but now you want to avoid it tipping too far towards knowledge.
Instead, the better metaphor is to think of knowledge as a pathway to skill. You teach knowledge, and skills are the end result.
Keir Bloomer, who I quoted at the start, appears to realise this.
“The problem is we did not make sufficiently clear that skills are the accumulation of knowledge. Without knowledge there can be no skills.”
This is absolutely the right way of thinking about it.
By contrast, I am wary when people say 'oh of course I am not anti-knowledge, we just need to teach both knowledge and skills'. That's not really how it works. The point is that you can't teach skills directly.
Vague curriculum & assessment statements
At best, the vague and woolly skills statements that make up CfE are true but useless. They will the ends without giving any guidance as to the means. They leave it up to teachers to do all the heavy lifting of specifying and resourcing content.
The Sunday Times article makes this argument. It says these kinds of statements are great for experienced teachers and well-resourced schools who could fill in the missing details and knowledge gaps that, but not for weaker teachers and schools who needed more guidance.
But there is an even more negative interpretation than this, which is that at worse, these kinds of vague statements actively prevent teachers designing activities that will help students learn.
That was my experience of teaching a similar skills-based curriculum in England (the 2007 version of the national curriculum). Even brilliant and experienced teachers could not make it work. That’s because it was based on the premise that there were no missing details and knowledge gaps for the teachers to fill in. Every time a teacher tried to fill in the gaps and teach a lesson focussing on times tables, or historical chronology, or the meaning of new words, inspectors and advisers would tell them that this was not in keeping with the curriculum.
These kinds of statements do not liberate teachers to teach their own knowledge. They force them into a straitjacket that says you don't need to teach knowledge, and that you can instead teach skills directly.
The further problem with these vague statements is that when they are used as an assessment tool, they are wildly unreliable. This has been known for a long time in the world of assessment. Even tightly defined statements like “can compare two fractions to identify which is larger” can be interpreted in ways that most students find trivially easy, and ways that they find impossibly hard.
If that is a problem with a tightly defined statement like this one, how much more of a problem will it be if you are being asked to assess a 500 word piece of writing against a series of statements like this one: “I can convey information, describe events, explain processes or concepts, and combine ideas in different ways.”
The Sunday Times quotes Keir Bloomer as saying that these statements are ‘childish rubbish’.
Further reading
I wrote about a number of these issues on my blog back in 2010-2014, when similar debates were happening in England. They might still be of interest to Scottish educators and those elsewhere having similar debates.
2011: Skills and Knowledge
2012: How grammatical knowledge underpins skilled communication
2013: Comparison of a knowledge-based chemistry curriculum with a skills-based one
2013: Do curriculum statements provide us with a shared language?
2013: The adverb problem
I also wrote two books that are relevant to these debates. Seven Myths about Education (2013) explains more about the importance of knowledge. Making Good Progress (2017) unpicks the problems that vague curriculum descriptors cause for assessment.
Improving assessment
I now work for No More Marking, a provider of online assessments.
Our Comparative Judgement assessments let you assess the "end goals" - complex skills like writing and mathematical problem solving. They do so in a way that removes the need for traditional moderation sessions and vague rubrics and curriculum statements.
Our multiple-choice assessments allow you to assess the "steps on the way" - the underlying components of knowledge that will allow students to acquire skill.
We hope these assessments are better aligned with the research evidence and in particular that they address the problem of assessing using vague curriculum statements.
This is a very interesting reframing of the knowledge-skills 'pendulum', thank you. The idea that factual knowledge underpins and lays the foundation for later skill development is definitely something I have been keen to explore more. I’d be interested to hear your thoughts on whether you think all skills necessarily depend on a foundation of explicitly taught facts.
For example, when learning to drive, the skill seems to develop largely through the act of driving itself, rather than through learning isolated facts about how each part of the car functions.
I wonder if this touches on a slightly different distinction, between propositional (factual) knowledge and procedural (ability based) knowledge.
Would you see these as fundamentally linked, or are there cases where one can develop more independently of the other? Many thanks again
I’m finding this “debate” is coming up a lot in the discussions about the use of LLMs in education. A lot of people who are coming to my school to talk about these things, or books/articles I am reading referencing how they are going to change education (they will) seem to suggest that students will be able to access, format and essentially download knowledge so instantaneously via LLMs and such like, that our teaching will shift dramatically to more “skills based” tasks. I have found this borderline impossible to reason with as the skills that are being suggested we pivot to in the face of LLMs aren’t really possible without the students having at the very least a foundational knowledge level in the first place.
How can a student pick apart a piece of writing about a historical event that they haven’t produced themselves, without knowing about the events first, or having a strong enough level of vocabulary to examine it? The better they know the topic, the better their examination of it will be. These things take time to embed and there are few shortcuts to this, and I’m not sure that outsourcing the knowledge gain is especially possible or conducive to developing the skills we want our students to develop. LLMs will doubtlessly have a place in our education system - we’d be foolish not to use them - but I can’t help but think if we expect them to learn knowledge on behalf of the students that we will only serve to flatten the intellect of our students to society’s detriment.