From human artefact to machine output: Automating the "art" of psychological measurement.
Product Description:
This study introduces an enhanced Psychometric Item Generator framework using large language models to automate psychological test development. Through three distinct trials, models generated high-quality items for AI trust, anxiety, and adoption scales. A validation trial involving human participants confirmed strong linguistic clarity, internal consistency, and clear factor structure. The method provides an accessible, cost-effective tool, though human oversight remains essential.
Keyword(s):
Large language models, psychometrics, Artificial intelligence
Product/Publication Type(s):
Peer-reviewed publications in scholarly journals Published/In Press
Target Audience:
Professionals, Students
Alternative Format:
Electronic (disc, CD, 508 compliant web posting)
To Obtain Copies (URL or Email):
COVID-19 Related Data:
N/A