IC2S2 tutorial explores how LLMs can simulate survey responses

FellMentKE · x · 2026-07-29

At IC2S2, the authors are running a tutorial on simulating human survey responses with large language models. The tutorial focuses on survey pretesting, hybrid human-and-simulated respondent designs, and missing-data imputation.

It emphasizes methodological rigor: participants are asked to treat LLM outputs as sensitive to persona construction and prompting choices, not as fixed samples. The tutorial introduces the QSTN framework to organize simulations and compare design choices using multiple evaluation metrics, with guided Python exercises and a discussion of the pitfalls of “silicon samples.”

Related event: IC2S2 Tutorial Explores Simulating Human Survey Responses with LLMs(2 posts)→

Original post →

More from Research

Research channel →