中国外语教育研究中心2013春季系列讲座之四：Growth Mixture Models (GMM) for Longitudinal Data Analysis
2013/06/18
讲座人：陈琦 博士（University of North Texas）
时 间： 2013年6月24日（周一）16:10-18:00
地 点： 北京外国语大学东院 逸夫楼202教室
讲座内容：
Growth Mixture Modeling (GMM) is a person-centered approach for analyzing longitudinal data. Using GMM, we can group individuals who are more similar to each other into categories. In this presentation, I will first introduce some concepts related to GMM, then present a simulation study examining the impact of ignoring a level of nesting structure in multilevel growth mixture models, and an empirical study applying GMM to investigate the differential effect of grade retention on the development of academic achievement from grade one to five on children retained in first grade over six years. Finally, I will briefly talk about the future directions of my research.
When applying GMM, researchers may assume that the higher level (non-repeated measure) units (e.g., students) are independent from each other even though it may not always be true. In the simulation study, three-level longitudinal clustered data were generated and then analyzed with the correct three-level model and the incorrect model that ignored the highest level of nesting structure separately. The simulation results showed, although over 90% of the replications resulted in the correct class solution under both true and misspecified models, ignoring a higher level nesting structure could still result in lower classification accuracy, overestimated lower-level variance components, and biased standard errors. The biased standard errors in turn affect the tests of significance for the fixed effects.
中国外语教育研究中心办公室
2013.6.17
版权所有 © 2006-2022 外语教学与研究出版社 高等英语教学网 京公网安备：11010802020838号 京ICP备18030989号-2