It's not often that equating (psychometric adjustment to put multiple test forms on a common measurement scale) comes up in the news. Unfortunately, it's completely misunderstood here. nypost.com/2025/08/16/u...

Ever wanted to investigate DIF by more than one variable at once? My colleague Ethan McCormick and I have created an NCME ITEMS module on DIF analysis with multiple background variables using moderated nonlinear factor analysis. Take a look! www.ncme.org/itemsportal/...

I taught an R for Data Science course last term. We work through examples in the r4ds book and then for homework students do problems I’ve created. A student asked if they can ask ChatGPT etc for help. I said, yes?

Saturday Night (Netflix) is pretty good. It feels like one big frenetic sketch, obviously exaggerating the mayhem and tension of down to the wire planning, but it feels very 70s and is cool to learn some of the history.

I finished the book. Au makes a good point about the circularity of differential item functioning (DIF) analysis, our main statistical tool for flagging biased test content, but ignores everything else we do to mitigate bias.🧵

I'm reading Wayne Au's (2023) Unequal by Design: High Stakes Testing and the Standardization of Inequality, a 140-page overview of how our capitalist education system in the US perpetuates inequities, using testing to turn students into commodities.

Mom and dad can hardly wait for school to start again. So true.

On my reading list this week is an overview of evidence-based standard setting by McClarty et al. It's a newish method that - I think - adds a bunch more impact data to panel discussions of where to set cut scores. journals.sagepub.com/doi/10.3102/...

Earthquake in norcal…

Interested in measurement? Do you like geeking out? Do you like geeking out with a good beer in beautiful Boulder? Then come to www.iomw.net