Improving Teaching: Chalk and Change podcast with Harry Fletcher-Wood
Sep 26, 2026 · 25 min · 12 segments
In this episode, we speak again to John Jerrim. He is Professor of Education and Social Statistics and Director of the Quantitative Social Science…
John JerrimGuest
Harry Fletcher-WoodHost
Actually, no, I do know on the first one, a really interesting one using PISA data, where we basically look at...

The change or the apparent change in kids skipping school over time since the pandemic.

Why I'm raising that one is there's a really interesting slash nerdy methodological thing in there, which changed only in a few countries.

And actually what you see very clearly in that data, which wasn't really documented anywhere, it took real detective work, was it had a big impact on the results to the point that there were some people that went and done some analysis and got it completely wrong, including some of the OECD, because they hadn't spotted it.

Also, we ended up using large language models, speaking to your second point to say, how Hey, chat GBT, do you notice any problem here? And we couldn't do it to find the problem that we had spotted.

So that was actually one of the most interesting ones that we've done over recent periods.

generally, I think it's really interesting because over the last few years, everyone started to realize What can I use this for and what can't I use it for and how I found it really useful recently is checking stuff.

The stuff that I do, does this make sense? What edits would you make here? Then I go back and think, shall I make that edit or not? Some people I know are using it to do coding and using codecs and crawls for their coding or whatever.

I'm not quite there yet because I don't have the, quite the trust yet to go out and farm it out to do it.

Stuff that would take me 10, 20 lines of code, it comes back with five or six pages.

I haven't got quite there yet, but definitely it's made me more productive, but also more accurate as well.
It's interesting the capacity that it has to proofread something with a degree of patience and attention that no human can achieve.
But like you, I've been using it to get feedback on bits and pieces of work and send it to writing.

Actually, no, I do know on the first one, a really interesting one using PISA data, where we basically look at...

The change or the apparent change in kids skipping school over time since the pandemic.

Why I'm raising that one is there's a really interesting slash nerdy methodological thing in there, which changed only in a few countries.

And actually what you see very clearly in that data, which wasn't really documented anywhere, it took real detective work, was it had a big impact on the results to the point that there were some people that went and done some analysis and got it completely wrong, including some of the OECD, because they hadn't spotted it.

Also, we ended up using large language models, speaking to your second point to say, how Hey, chat GBT, do you notice any problem here? And we couldn't do it to find the problem that we had spotted.

So that was actually one of the most interesting ones that we've done over recent periods.

generally, I think it's really interesting because over the last few years, everyone started to realize What can I use this for and what can't I use it for and how I found it really useful recently is checking stuff.

The stuff that I do, does this make sense? What edits would you make here? Then I go back and think, shall I make that edit or not? Some people I know are using it to do coding and using codecs and crawls for their coding or whatever.

I'm not quite there yet because I don't have the, quite the trust yet to go out and farm it out to do it.

Stuff that would take me 10, 20 lines of code, it comes back with five or six pages.

I haven't got quite there yet, but definitely it's made me more productive, but also more accurate as well.
It's interesting the capacity that it has to proofread something with a degree of patience and attention that no human can achieve.
But like you, I've been using it to get feedback on bits and pieces of work and send it to writing.
The rest of this transcript — segmented and speaker-labeled, so you land on the exact moment something was said
Search every transcript — by keyword, by phrase, or by meaning, across every show Radar indexes
Trends — what is surging across podcasts, measured against its own baseline
Alerts — when a name you follow appears in a newly indexed episode
No account is needed to search Radar.