3-Read Friday #103

Checking for understanding, tutor-proof tests, and AI localisation

Here are three blog posts that I found interesting this week.

1. Strat Chat - Checking for Understanding by Shaun Brien

I love the format of this post! Shaun argues that checking for understanding is not about gauging confidence or asking if students "get it," but about using structured, whole-class strategies like TAPPLE to make thinking visible, catch misconceptions in real time, and inform responsive teaching decisions.

2. Is it possible to develop a tutor-proof test? by Daisy Christodoulou

This Queen of Assessment is back with a belter. Daisy makes the case that the concept of a "tutor-proof" selection test is fundamentally flawed because the practice effect ensures any test format becomes coachable. Moreover, replacing curriculum-based assessments with abstract reasoning tasks risks equal social exclusion while promoting less educationally valuable preparation.

3. Context counts: Measuring how AI reflects local realities in education by AI for Education

One of our Eedi Learning Design team, Claire, shared this paper with me. The paper shows that AI models perform significantly worse at contextualising educational content for low- and middle-income countries, with existing benchmarks revealing substantial gaps in cultural accuracy, geographic coverage, and the ability to move beyond stereotypes. This means that teachers and policymakers in these contexts must apply careful human review to any AI-generated materials.

Have a great weekend!

Craig

🏃🏻‍♂️Before you go, have you… 🏃🏻‍♂️

… checked out my brand-new book series: The Tips for Teachers guides to…

And checked out my all-new, ad-free, Mr Barton Maths website