各位老师，您好：
今天下午1:30在小楼三楼大会议室举办的报告信息如下。如果您感兴趣，欢迎您去听。
报告1：计量心理学的发展和前瞻 ：程莹 圣母大学心理系
程莹博士, 2003年毕业于中国科学技术大学，并与同年秋赴美留学, 师从教育和心理测量学专家张华华教授, 开始项目反应理论及其实践应用的研究。2007年获美国伊力诺依大学香槟分校统计硕士学位, 2008年获该校计量心理学博士学位。毕业后任教于美国圣母大学心理系，并创建心理与教育测量实验室，现有博士生三名。自2012年7月起担任圣母大学教育研究所Fellow。其研究成果自2007年以来被Psychometrika, Applied Psychological Measurement British Journal of Mathematical and Statistical Psychology, Behavioral Research Methods 等业内重要研究期刊刊载，并于2010和2011年两度应邀参与美国自然科学基金会的项目评审工作。2009年获得美国教育测量协会Bradley Hanson 教育测量研究贡献奖, 2012年作为唯一获奖者获该协会突出青年学者奖。
报告摘要：
计量心理学是一个新兴的交叉学科，其主要分支包括心理学与教育学和统计科学的交叉 Psychometrics)，认知心理学与数学建模的交叉 (Computational Modeling and Mathematical Psychology)，以及心理学与决策科学的交叉(Judgment and Decision Making) 等等。这场报告将主要介绍Psychometrics （心理测量测度）这个分支的发展和现状，代表性的计量心理学重镇，并在此基础上探讨该方向发展的前瞻。
报告2：Latent Structure Models vs. Diagnostic Classification Models: What’s new? Matthias von Davier Research & Development Division at Educational Testing Service
Dr. Matthias von Davier is a research director in the Research & Development Division at Educational Testing Service (ETS) and Honorary Senior Research Fellow at Oxford University in the United Kingdom. He joined ETS, which is located in Princeton, NJ, USA, in 2000. He earned his Ph.D. in psychology from University of Kiel, Germany, in 1996, specializing in psychometrics. At ETS, Dr. von Davier manages a group of researchers concerned with methodological questions arising in large-scale international comparative studies in education such as PISA, TIMSS and PIRLS. He is the winner of the 2006 ETS research scientist award and the 2012 NCME Bradley Hanson Award for contributions to educational measurement. Prior to joining ETS, Dr. von Davier was co-director of the “Computer as a tool for learning” section at the Institute for Science Education (IPN) in Kiel, Germany, and was an associate member of the Psychometrics & Methodology Department of IPN. His research interests involve extensions of item response theory, diagnostic classification models, latent class analysis, mixture distribution models as well as computational statistics and model-data fit. Recently, he was elected to be the next editor of the British Journal of Mathematical and Statistical Psychology.
Abstract Models for categorical observed variables with nominal latent variables have been referred to as latent structure or latent class models. Diagnostic classification models have been introduced in recent years to enable estimation of multiple discrete skills based on a number of observed categorical variables. The promise of these types of models is that more than just a single proficiency score can be derived and students as well as educators can obtain more detailed information about the strengths and weaknesses of test takers. The talk will present some results that relate diagnostic models such as the DINA model to response error models on the one hand and more general latent variable models on the other hand. Some new results on model equivalencies and identifiability will be presented and potential applications will be illustrated.
此致敬礼！
认知神经科学与学习国家重点实验室
2012年9月25日
