England says it has high standards and a knowledge-rich curriculum. So how can you pass GCSE knowing 20%? On standards, attainment and the curiosity gap.
Thanks Chris, that's really helpful to see. Do you think that the causality could potentiallt be working the other way around (low attainment in school making students feel that, even though they are curious generally, their curiosity doesn't translate to results in school, which pushes down their love of school score)?
Great point, Max, and it’s one PISA can't settle, since it's a single snapshot, so direction of causation is out of reach. If I was forced to put money on it, I'd bet that it runs both ways: low curiosity dragging down attainment, but also the mechanism you describe - repeated experience of your curiosity not 'paying off' in grades/attainment slowly affecting how you feel about learning in school. The motivation literature (esp around academic self-concept) backs up your version and it rings true in my experience, too. I've taught plenty of kids who are obviously curious when you chat to them outside lessons but the moment we hit the subject they’re struggling in, it vanishes.
Thanks Chris. Found this article really informative. For your analysis on the curiosity gap of PISA countries, were you able to see anything meaningful in terms of an average trend line? It looks like there are some countries on the left hand side with a relatively high curiosity gap that seem to have lower attainment gaps.
Is the trend meaningful? I’ve put the stats on there now (and probably should have originally) [note - I can’t attach a pic apparently, so will amend my post…]:
> The p value means it’s statistically significant.
> The R^2 means attainment gap explains about 9% of the variation in curiosity ‘gap of gaps’ (i.e. between general and school curiosity in poorer and wealthier students).
> The r (shown as R on the graph) is the correlation strength, which is probably the most important: about 0.3 – on the cusp of small and medium. Which isn’t huge, but education trends generally aren’t. And it would be larger (about 0.38) without the outliers. Talking of which…
I’ve added the two top left – Cambodia and Uzbekistan. Both are low-performing systems (averaging about 340-350, near the PISA floor), so there's little room for richer pupils to pull ahead. In terms of whether their education policies are having an effect… would be great to know, but I have to confess I don’t know much about them or whether they have anything in common except having a low attainment gap, being low attaining, and having a positive curiosity ‘gap of gaps’ (unlike many low-performing education systems)
The argument- you should learn this because when “we” understand more about atoms we will have nuclear fusion and end climate change and species extinction, seems a bit tenuous.
A better argument for a general knowledge base is if you have it you will be able to ask questions when someone tells you something to see how much you should base your choices on what they are saying.
Utilitarian arguments for most school physics topics won’t stand up to a check of how many people will use it - ask how many people use their knowledge of say Boyle’s law or how many people are going to dedicate a hard 10-20 years to make a contribution to quantum physics and it won’t make sense for most people to put any effort into these in school.
My argument that it’s valuable for students to know more about atoms so that ‘we’ can one day get nuclear fusion was the best I could do on the spot when I had a couple of weeks to get a struggling class through their exams. It was not the launch of a manifesto, but it did engage the student quite a bit more than the revision did!
I understand the powerful knowledge arguments – expert guidance, learning important habits of mind, the only opportunity you’ll have to learn it, the equity perspective on who gets to learn it – but I do have days when I wonder how powerful e.g. Boyle’s law is for most students, if they still don’t get it properly after a couple of hours going over it, they’ll probably forget it soon after school and never use it again – unless they become a science teacher. Extended discussion of that though will have to wait for a full post.
Overall thesis: there are many reasons to teach science (of which the utilitarian view is one, but less significant than others, e.g. your claim that we should be able to have an informed take on evidence is probably more fundamental). We should put these forward, transparently, to students, otherwise we often end up justifying the curriculum by the end point – the exam.
It’s interesting that you had to come up with that on the spot. That is - no part of lesson planning or curriculum exposition includes documenting the point learning something.;
That people don’t object to this state more suggests most people are quite incurious about what is worth teaching everyone.
Even if someone has a vague - gas laws are easy to reproduce and link algebra and physical data, they were historically important or are key to designing efficient engines and heat and cooling systems it’s not obvious where the point of diminishing returns is for a high school student. Perhaps covering one in-depth would be a better use of time for most people and for the rest just mention there are others you can learn about if you are inclined.
My point is it is extremely hard to justify any one item based on any criteria other than someone put it in the curriculum and they did not explain themselves but your exam results will matter to you.
The point of that lesson was revision and passing exams. To the student – and to me, when I thought about it – that didn’t feel like enough.
‘That people don't object to this more suggests most people are quite incurious about what is worth teaching everyone.’ Exactly. Digging into that is the whole project.
On diminishing returns – agreed, and we barely have this conversation. We teach the future research scientist the same way as the student who’ll only ever need to reason about science in the news. Is that the best way to go about it?
We justify the curriculum by saying an expert deemed it important – which is true – and that there’s an exam on it. Not enough. If someone decided these hours were worth spending, the least we owe students is an explanation of why.
There’s an irony here. Being good at studying for and taking exams is one of the most useful things you can learn. Want to be a police officer - lots of exams, run a restaurant kitchen- exams, sports coach - exams.
Being able to assess what will be covered and how best to prepare and then write an exam is one of the most useful skills anyone can have.
Knowing what it would take you to go from where you are to where you need to be is key to minimizing stress for the rest of your life.
I see what you're saying, Stan - I'm not denying good exam-takers exist, I've envied them, and it's useful in life.
But I'd separate two things. There's the metacognition: working out what's required, gauging your own understanding, closing the gap. Genuinely useful, but it goes beyond exams. Start a new job and you do exactly that, with no exam in sight. So the transferable skill isn't exam-taking; it's knowing how to learn, which happens to be particularly useful when it comes to exams.
And then there’s the debate about whether you can teach ‘being good at exams’ at all, separate from the content. Many would say you can't – it’s predicated on mastering the domain knowledge. So, what matters isn’t necessarily being able to take the exam but being able to master a body of knowledge (which the exam just tests). Competence is what's behind it, and that's the bit I think is worth justifying to students.
Exams are simply a cheap way to measure domain knowledge and domain competence.
People who argue exams won’t tell you everything about mastery of a domain would need to show an exam is not just imperfect but worse than something equally available.
People who argue you can’t learn how to learn domain knowledge have to explain why medical students turn to tools like Anki for spaced repetition practise.
It’s obvious to most people who want to retain knowledge that there are good and bad approaches and gaining confidence in using the good ones will remove a massive amount of stress and effort.
Some people object that memorizing is not understanding and some students may focus on memorizing alone. They need to explain how that is relevant because you cannot think about things you do not remember.
Thanks Chris, that's really helpful to see. Do you think that the causality could potentiallt be working the other way around (low attainment in school making students feel that, even though they are curious generally, their curiosity doesn't translate to results in school, which pushes down their love of school score)?
Great point, Max, and it’s one PISA can't settle, since it's a single snapshot, so direction of causation is out of reach. If I was forced to put money on it, I'd bet that it runs both ways: low curiosity dragging down attainment, but also the mechanism you describe - repeated experience of your curiosity not 'paying off' in grades/attainment slowly affecting how you feel about learning in school. The motivation literature (esp around academic self-concept) backs up your version and it rings true in my experience, too. I've taught plenty of kids who are obviously curious when you chat to them outside lessons but the moment we hit the subject they’re struggling in, it vanishes.
Thanks Chris. Found this article really informative. For your analysis on the curiosity gap of PISA countries, were you able to see anything meaningful in terms of an average trend line? It looks like there are some countries on the left hand side with a relatively high curiosity gap that seem to have lower attainment gaps.
Thanks Max, and great questions.
Is the trend meaningful? I’ve put the stats on there now (and probably should have originally) [note - I can’t attach a pic apparently, so will amend my post…]:
> The p value means it’s statistically significant.
> The R^2 means attainment gap explains about 9% of the variation in curiosity ‘gap of gaps’ (i.e. between general and school curiosity in poorer and wealthier students).
> The r (shown as R on the graph) is the correlation strength, which is probably the most important: about 0.3 – on the cusp of small and medium. Which isn’t huge, but education trends generally aren’t. And it would be larger (about 0.38) without the outliers. Talking of which…
I’ve added the two top left – Cambodia and Uzbekistan. Both are low-performing systems (averaging about 340-350, near the PISA floor), so there's little room for richer pupils to pull ahead. In terms of whether their education policies are having an effect… would be great to know, but I have to confess I don’t know much about them or whether they have anything in common except having a low attainment gap, being low attaining, and having a positive curiosity ‘gap of gaps’ (unlike many low-performing education systems)
The argument- you should learn this because when “we” understand more about atoms we will have nuclear fusion and end climate change and species extinction, seems a bit tenuous.
A better argument for a general knowledge base is if you have it you will be able to ask questions when someone tells you something to see how much you should base your choices on what they are saying.
Utilitarian arguments for most school physics topics won’t stand up to a check of how many people will use it - ask how many people use their knowledge of say Boyle’s law or how many people are going to dedicate a hard 10-20 years to make a contribution to quantum physics and it won’t make sense for most people to put any effort into these in school.
My argument that it’s valuable for students to know more about atoms so that ‘we’ can one day get nuclear fusion was the best I could do on the spot when I had a couple of weeks to get a struggling class through their exams. It was not the launch of a manifesto, but it did engage the student quite a bit more than the revision did!
I understand the powerful knowledge arguments – expert guidance, learning important habits of mind, the only opportunity you’ll have to learn it, the equity perspective on who gets to learn it – but I do have days when I wonder how powerful e.g. Boyle’s law is for most students, if they still don’t get it properly after a couple of hours going over it, they’ll probably forget it soon after school and never use it again – unless they become a science teacher. Extended discussion of that though will have to wait for a full post.
Overall thesis: there are many reasons to teach science (of which the utilitarian view is one, but less significant than others, e.g. your claim that we should be able to have an informed take on evidence is probably more fundamental). We should put these forward, transparently, to students, otherwise we often end up justifying the curriculum by the end point – the exam.
It’s interesting that you had to come up with that on the spot. That is - no part of lesson planning or curriculum exposition includes documenting the point learning something.;
That people don’t object to this state more suggests most people are quite incurious about what is worth teaching everyone.
Even if someone has a vague - gas laws are easy to reproduce and link algebra and physical data, they were historically important or are key to designing efficient engines and heat and cooling systems it’s not obvious where the point of diminishing returns is for a high school student. Perhaps covering one in-depth would be a better use of time for most people and for the rest just mention there are others you can learn about if you are inclined.
My point is it is extremely hard to justify any one item based on any criteria other than someone put it in the curriculum and they did not explain themselves but your exam results will matter to you.
The point of that lesson was revision and passing exams. To the student – and to me, when I thought about it – that didn’t feel like enough.
‘That people don't object to this more suggests most people are quite incurious about what is worth teaching everyone.’ Exactly. Digging into that is the whole project.
On diminishing returns – agreed, and we barely have this conversation. We teach the future research scientist the same way as the student who’ll only ever need to reason about science in the news. Is that the best way to go about it?
We justify the curriculum by saying an expert deemed it important – which is true – and that there’s an exam on it. Not enough. If someone decided these hours were worth spending, the least we owe students is an explanation of why.
There’s an irony here. Being good at studying for and taking exams is one of the most useful things you can learn. Want to be a police officer - lots of exams, run a restaurant kitchen- exams, sports coach - exams.
Being able to assess what will be covered and how best to prepare and then write an exam is one of the most useful skills anyone can have.
Knowing what it would take you to go from where you are to where you need to be is key to minimizing stress for the rest of your life.
I see what you're saying, Stan - I'm not denying good exam-takers exist, I've envied them, and it's useful in life.
But I'd separate two things. There's the metacognition: working out what's required, gauging your own understanding, closing the gap. Genuinely useful, but it goes beyond exams. Start a new job and you do exactly that, with no exam in sight. So the transferable skill isn't exam-taking; it's knowing how to learn, which happens to be particularly useful when it comes to exams.
And then there’s the debate about whether you can teach ‘being good at exams’ at all, separate from the content. Many would say you can't – it’s predicated on mastering the domain knowledge. So, what matters isn’t necessarily being able to take the exam but being able to master a body of knowledge (which the exam just tests). Competence is what's behind it, and that's the bit I think is worth justifying to students.
Exams are simply a cheap way to measure domain knowledge and domain competence.
People who argue exams won’t tell you everything about mastery of a domain would need to show an exam is not just imperfect but worse than something equally available.
People who argue you can’t learn how to learn domain knowledge have to explain why medical students turn to tools like Anki for spaced repetition practise.
It’s obvious to most people who want to retain knowledge that there are good and bad approaches and gaining confidence in using the good ones will remove a massive amount of stress and effort.
Some people object that memorizing is not understanding and some students may focus on memorizing alone. They need to explain how that is relevant because you cannot think about things you do not remember.