A comparison of 20 heterogeneity variance estimators in statistical synthesis of results from studies: a simulation study

• Published in: Statistics in medicine Vol. 36; no. 27; pp. 4266 - 4280
• Main Authors: Petropoulou, Maria, Mavridis, Dimitris
• Format: Journal Article
• Language: English
• Published: England Wiley Subscription Services, Inc 30.11.2017
• Subjects: Computer Simulation; coverage probability; Data Interpretation, Statistical; heterogeneity variance estimators; Humans; length of confidence interval; Markov Chains; mean absolute estimation error; Medical statistics; Meta-analysis; Meta-Analysis as Topic; Monte Carlo Method; Simulation; simulation study; Statistics as Topic; coverage probability; mean absolute estimation error; heterogeneity variance estimators; simulation study; length of confidence interval

When we synthesize research findings via meta‐analysis, it is common to assume that the true underlying effect differs across studies. Total variability consists of the within‐study and between‐study variances (heterogeneity). There have been established measures, such as I2, to quantify the proport...