报告主题：Application of MCMC algorithm with Davidian curves in IRT models

报 告 人：张 雪 副教授

报告时间：2024年12月7日（周六）上午10：00-12：00

报告地点：腾讯会议（会议号：211-580-463）

报告摘要: Normality of latent traits is a routine assumption made when estimating item parameters for item response theory (IRT) models, but it might be unrealistic with some datasets. The purpose of this research was to present two novel Markov chain Monte Carlo (MCMC) method for ordinal items with flexible latent trait distributions (i.e., normal, skewed, and bimodal). Specifically, the Davidian curve (DC) was used to approximate the distribution of latent traits. The performances of the proposed methods were illustrated via simulation studies and a real data example. Preliminary results indicated that the proposed methods could fit normal and bimodal distributions well and skewed distributions reasonably well, and the method provided good estimates of item parameters.

报告人简介：张雪，东北师范大学中国农村教育发展研究院副教授、统计学博士，美国明尼苏达大学访问学者。主要研究方向为项目反应理论、认知诊断模型的统计推断及应用。在MBR、BJMSP、APM、JEM等期刊发表多篇学术论文。主持教育部人文社会科学研究项目青年基金项目1项、国家自然科学基金青年基金项目1项，参与多项高层次教育评估项目。