Psychometrics and Linguistics in ChatGPT-Generated Reading Tests Compared with CET-4

Quick email FYI research alert post.  Not open access.

 

“The findings corroborate the conclusion that ChatGPT may function effectively as a supplementary tool in low-stakes assessment; however, substantial refinements in item quality are imperative before its application in high-stakes testing.”

 

Psychometrics and Linguistics in ChatGPT-Generated Reading Tests Compared with CET-4 - Bingwei Li, Qixiu Qin, Wanqing Li, Chunhua Mao, 2026 
https://journals.sagepub.com/doi/10.1177/07342829261444828?_gl=1*1p1wlt1*_up*MQ..*_ga*MTYxOTA1NjY2Ny4xNzc3Mzg0MTgy*_ga_60R758KFDG*czE3NzczODQxODEkbzEkZzAkdDE3NzczODQxODEkajYwJGwwJGg0NDg5MzkwNDI.

Pardon typos and spelling errors-Message may be sent from iPhone and I've always had spelling problems :)