每周论坛之五(总第169期):柯紫筠(中山大学),2018年4月2日下午

Reliability Matters in Meta-Analysis: A Better Approach to Correcting for Unreliability

发布人:高级管理员 发布日期:2018-04-02
主题
Reliability Matters in Meta-Analysis: A Better Approach to Correcting for Unreliability
活动时间
-
活动地址
中山大学东校园心理学系(南学院楼C座)305
主讲人
柯紫筠博士

报告人介绍:

讲座介绍:

每周论坛之五(总第169期)

报告人:柯紫筠博士

报告人单位:中山大学心理学系

题目:Reliability Matters in Meta-Analysis: A Better Approach to Correcting for Unreliability

时间:2017 4  2   (周一,14:20-16:00)

地点:中山大学东校园心理学系(南学院楼C座)305

报告简介

As a powerful tool for synthesizing information from multiple studies, meta-analysis has enjoyed high popularity in many disciplines. Conclusions stemming from meta-analyses are often used to direct theory development, calibrate sample size planning, and guide decision making or policy making. However, meta-analyses can be conflicted, misleading, and irreproducible. One of the reasons for meta-analyses to be misleading is the improper handling of unreliability. We showed via a simulation study that current meta-analysis procedures with or without corrections for unreliability frequently detected nonexistent effects, and provided severely biased estimates and interval estimates with coverage rates far below the intended level. A better approach to correcting for unreliability was proposed and evaluated via a simulation study.

报告人简介

柯紫筠,女,中山大学心理学系讲师。研究兴趣为元分析、结构方程模型元分析、增长曲线模型分析、时间序列分析、稳健分析等。